Hacker News Reader

Comment by ppeetteerr | Hacker News Reader

My question is if your conclusion is that blockchain is a solution in search of a problem, what was all that effort being spent on (not just in finance, but everywhere blockchain is being considered)?

I'm not saying that blockchain is not a valid solution to a problem, but as a veteran of the online industry, I can say I've seen a lot of hype around new technology, actual effort being put into it, and then the realization that the effort was for nothing. Either the solution is no better than the old solution, just different, or the solution is so obscure, that the cost of maintaining it is greater than the value it generates.

In the end, the costs of the choices of these institutions are being paid for by you and I, and I hope they don't go too deep into their project before realizing: “Basically, it became a solution in search of a problem,”

View on HN

> what was all that effort being spent on

Not being left behind if you're wrong.

I'm in a tech-adjacent industry and see it all the time. Competitor A comes out with "shiny new technology" and markets the hell out of it. My company panics because we're suddenly lagging behind and rushes a competing project into development. Competitor B sees all of this and hops in too. All of a sudden it's the new direction of the industry and dominating trade shows, despite the origin being a mediocre product with some clever marketing. 5 years later it's never spoken of again.

I'm not familiar enough to say for sure this is what happened with "blockchain", but from a distance it has a lot of familiar markers.

An example from the top of your head ?

GPU acceleration in scientific computing. My god in 2011 it was the only thing anyone talked about. Every COTS solver package cobbled together support in less than a year and were promising order of magnitude speedups.

In 2018 it's still hanging around but realizing a 20% speedup on certain classes of problems. Maybe.

Amdahl's law strikes again.

Didn't that tech shifted to AI/ML? Isn't that considered scientific computing by some measure?

I'm definitely not an expert so I'm maybe missing something here.

"Scientific Computing" was probably too general of a term to use. I focus more on FEA/CFD type problems where it's effectively been a bust. Most of our benchmarks show 8 cpu cores + 1 GPU is roughly equivalent to 10 cpu cores, and a second GPU has little additional benefit.

I'm sure there are other disciplines that have seen a larger benefit.

> GPU acceleration in scientific computing

> it's still hanging around but realizing a 20% speedup on certain classes of problems

It really varies though. In grad school I felt like god when some molecular dynamics simulations running on my pair of GTX 760s finished faster than the same thing running on my advisor's 96-core CPU cluster.

Point taken, though, that it was over-hyped. There were lots of conference presentations with GPU-acceleration in the title (kind of like deep learning is now).

Wasn't your advisor just using a smaller, more conservative value for h?

Relevant war story: I was one of the engineers who initially started looking at GPGPU computing at one of my former employers. This was circa 2007 and people were getting really excited thanks largely to NVIDIA's hype machine. 2007 BTW, was when the first Teslas were making their way into the wild.

I was part of a group that developed signal processing algorithms and my boss's boss pulled me aside and asked me what we ought to do about this stuff. I told him that there are a few applications where this works very well, but it's pretty niche overall and we shouldn't do more than what we'd already done. He was a bit surprised because he thought I'd be one of the kool-aid drinkers given how much I knew about and liked playing with the tech. But he deferred to my judgement.

It turned out to be one of the most significant conversations I've had. I was still a junior engineer fresh out of college, but my bosses realized that I could be trusted put aside naive enthusiasm about the tech and make cool-headed decisions about what was strategically important. And I learned to trust my intuition about what was useful and what was just hype.

At the same time, I really wish that GPUs would take off in neuroimaging, where 90% of the steps performed in an analysis pipeline are not modeling-related, but rather basic image manipulations / pre-processing (i.e., scaling, smoothing; stuff that GPUs are actually good at).

Semantic web related technologies. Oh man, this was the future rdf triplets, ontologies ...

"Digital Transformation"

that one is still hot, according to our marketing...

Well, "digital transformation" is not a technology -- it's this strategic paradigm of how technology is diffused across an organization. The opposite of "digital transformation" is "IT infrastructure" -- the idea that technology works as an infrastructural layer supporting business operations, rather than dispersed across operations themselves.

"IT" has for long been this back-office area where specialists get to dictate business policies because they know computers. (A classic on this genre is the Bastard Operator from Hell [0]) -- but now lusers bring their own computers and phones to work, and hell if they don't have technological ideas of their own. IT has been commoditized, cloudified, outsourced.

If "digital transformation" sounds bogus is because management consultants are bogus. But in practice it can be a real thing that's not as much about new sexy tech (in the sense of blockchains, RISC chips and so on) as about putting what's already available into operation using business knowledge.

[0] https://en.wikipedia.org/wiki/Bastard_Operator_From_Hell

X. M. L.

My major questions have always been: where is the business value in blockchain applications? what can a blockchain app do that can't be done by non-blockchain (beyond decentralization - bc I don't think this produces much business value)?

Sending money without fees and delay? Have you ever tried a normal bank transfer? How long did it take? Have you ever tried, say, nano? It literally takes seconds and 0 fees. Here is some value.

The delay and cost is a function of mandatory compliance and recourse, rather than anything to do with technology.

I’m glad some people get it. Banks do zero delay and zero cost transactions between themselves already. The idea that “if only we coukd figure out where to use blockchain, everything would be FREE” is really naive. You take away a banks ability to charge for X, don’t be surprised when Y gets a new fee.

As to decentralized, there seems to be a wonderful libritarian ideal around cryptocommodities but I’m reality they are much more centralized than anyone wants to admit. There are probably 10 people that if they wanted to could absolutely demolish bitcoin.

Too much hype around all of this. Not even getting into the massive waste of power and potential.

Just about everyone that's worked in finance professionally gets it - netting, collateral calls etc all happen daily and instantaneously and have been for decades.

Having worked in computational finance the entirety of my career, I still don't totally understand how this combination of Dunning-Kruger and Libertarian leanings somehow became a thing.

The financial times while back had an interesting take: there's a structural reason why VC throws money at seemingly non-nonsensical ideas (Theranos, wireless charging etc) - many have a policy portfolio mandate to shoot for 100x returns and as such they will throw money at impossible things as part of their fiduciary duty. So, especially in the climate of free money, funding a slightly-better-widget-startup takes a backseat to things with lottery odds that may not make any sense at all.

The power consumption is the most goofy part of the whole idea though.

It is the fatal flaw of POW. It is like just pretending that the power cost is not part of the cost of a transaction.

The delays have little to do with technology, and a lot to do with regulations and bureaucracy. Largely put in place to combat money laundering, fraud, terrorist financing, etc etc. Yes, there's some small banks that have horrible APIs and shoddy servers, but for most transfers the delays come down to things that if cryptocurrency were ever mainstreamed would be tacked onto it.

That is why it takes time to get approved on coinbase and other exchanges that try to follow the law.

There is literally no place to tack on a middleman with nano/btc. Yes it will take some rethinking of how we combat what society conceives as improper use of funds.

> There is literally no place to tack on a middleman with nano/btc.

BTC is just a widely distributed set of middlemen imposing delays and costs (both mining rewards and transaction fees are costs to everyone holding or using BTC that run from holders/users to miners.)

So the people I have to go through to convert my crypto to real money aren't middlemen?

I plan on only accepting crypto for consulting this year, it filters out clients and I have limited time.

At some point I will exchange to USD to pay whatever I owe in taxes and to pay any vendor that does not take crypto. I imagine if I were an unscrupulous actor I would skip the tax step and avoid vendors that don't accept crypto.

Good for you. I have rent to pay, so I need to be paid in USD. That brings us back to my original question: Are the people I have to go through to exchange crypto to USD not middlemen?

That was an example of how to avoid conversion to 'real money' through spending and accepting only crypto.

Why can't you build on top of these protocols, nothing is stopping them from being underlying technology that new banks and financial institutions can be built on top of.

Bitcoin has both fees and delays. You can see what they are now:

https://bitinfocharts.com/comparison/bitcoin-transactionfees...

https://blockchain.info/charts/avg-confirmation-time

Who knows what they would be if bitcoin was actually used to buy things.

He's not talking about Bitcoin, do people realize there's literally 1000+ projects (Most irrelevant) with some clear innovation going on. We're already way past bitcoin and PoW as the future of blockchain tech.

> Have you ever tried a normal bank transfer? How long did it take?

For less than $5,000 it is free and instantaneous in the United States (e.g. Zelle). For more, it is free and instantaneous (wire; my bank waves the fee; most charge $15 to $35).

The number one thing holding back the blockchain from becoming a proper database technology is its proponents insistence on applying it to problems that already work faster and cheaper.

Or Dwolla! Funny - searched google for Dwolla and Zelle came up as the first match

In the UK we can do that without the blockchain, in many situations, thanks to the real-time payment system [0]. So for me, a normal bank transfer is free and easy, whereas every time I've tried to do something involving blockchain it's meant delays, fees, and messing around on shady-looking websites.

I take the point that in other places the norm is a lot worse and blockchain systems are producing better results, but it won't be long before banks get their act together and improve their services to match modern expectations (for fear of becoming obsolete), and then the blockchain competitive advantage is gone.

[0] http://www.fasterpayments.org.uk/

Yeah, in Australia it's currently overnight between banks (but generally instant from one account to another within a bank), and they're bringing out a new real-time payments platform over the next few months so you will be able to transfer instantly to anybody knowing only their 'pay ID', which can be a mobile number, Australian Business Number (ABN), etc.

SEPA Instant Pay allows transactions of up to 15k in 10 seconds (bigger transactions fall back to the 1 day SEPA rule, I think), and some banks provide it for free. This is only relevant within the EU and associated regions, obviously.

In Canada, the gov'ment told the banks they had to offer cheap, instant transfers. Today, it's free to send money to anyone in the country and it takes minutes.

Nothing.

Decentralisation is the selling point of blockchain technology.

Anyone building on a blockchain that doesn't think decentralisation is beneficial in and of itself is a conman.

This is really the main selling point, decentralizing consensus so no one party becomes to big to fail, or so big they are unable to be opposed which leads to tyranny.

The sad part is that the blockchain is the least interesting part of this hype. The interesting part is making all mutations of state require signed inputs which has given a really needed push to the PKC UX. Metamask is fucking amazing and interacting with applications that utilize metamask as the source of identity is so simple, something that you'd be hard pressed to find before all this hype.

Atomic transactions of money on one hand, and ownership on the other.

For big amounts, transfers of money are either slow (big wire transfers between banks) or unsecure (cash, how do you know if those bills are real?). Therefore either the money or ownership stays insecurely in limbo on one side.

Therefore escrow services currently provide a solution for this, or notaries. But in the end they are still not atomic, and not free either.

If both money and proof of ownership are both on blockchain, you can do a REAL atomic transaction. This could work for selling cars, houses, etc, on an international level without much fuzz.

"My question is if this is your conclusion is that it's a solution in search of a problem, what was all that effort being spent on (not just in finance, but everywhere blockchain is being considered)?"

That strikes me as a very rich question that would be interesting to investigate. Yes, we know people can get involved in hype cycles, but what was it about this technology that made it the thing that got hyped exactly?

Perhaps a subconscious association with BitCoin and getting rich quick hit the people most prone to hype? Some of the blockchain projects being proposed, I honestly couldn't see how they were planning on profiting from it. I don't just mean profiting enough, but profiting at all; what was the mechanism by which money is transferred from some set of customers to the blockchain creators? Preferably with the transfer of some valuable good or service in the other direction, but, you know, technically that's optional from a business perspective.

I understand the answer for new cryptocurrencies; the creators start with more of it than the incoming users and that becomes valuable as the currency becomes valuable. The non-currencies I didn't always see an answer. Which is why my best guess is some sort of subconscious belief that anything that has a blockchain will automatically somehow become wildly valuable via (underpants gnomes step 2 here). It's kinda a copout answer but it's the best I've got. Improvements? (Again, beyond just "people do hype." Why this hype and not something else?)

I look at it differently:

Brick-and-mortar plus mail-order was a viable business model for decades. Building the network infrastructure and software interconnections to move from that state to today’s online commerce and delivery model took years to accomplish and was (is) filled with hyped innovations.

Blockchains may never reach their hyped potential, or we may simply be attempting to evaluate at very early stages vs. the point where utilities are clear.

Do you look at every single new technology this way? If so how do you find time for anything else? If not what makes blockchain different?

I’m not sure I follow. I did not mean to imply I’m studiously watching every permutation of the blockchain hype cycle, simply, when it comes to new technologies which are billed as successors to long-lived incumbencies, I tend to take the middle ground.

I try to keep a measured view of the optimisic hype and the pessimistic nay-saying, and don’t expect clarity for a few more years regarding blockchains’ potential.

In general this philosophy towards technology has served me well enough. I miss out on the bleeding edge, but don’t fall into the trap of being “all in” before realizing a concept’s promise was oversold.

Asymmetry: it is very hard to irrefutably defeat someone proposing some highly amazing, low-probability-of-coming-to-pass future. The best you can do is say "hard to see how that happens, but who knows?..." Meanwhile the bullshitter is utterly convinced by their own koolaid. In the end it is a bug in the human brain firmware.

Reply on news.ycombinator.com